Paclitaxel, Carboplatin, and Radiation Therapy in Treating Patients Who Are Undergoing Surgery for Stage III Non-Small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Paclitaxel, Carboplatin, and Radiation Therapy in Treating Patients Who Are Undergoing Surgery for Stage III Non-Small Cell Lung Cancer (NCT00096226), a na phase 2 study. The primary outcome is Mediastinal Nodal Clearance Rate.

2. Clinical Question

This distinction between the clinical question and the design rule is important. The registry describes what the investigators planned to assess, and the threshold pattern used to guide a conclusion. It does not report whether either criterion was met. The study therefore cannot be read from this record as demonstrating a particular observed clearance rate.

3. Trial Design

This was a phase II, single-group trial sponsored by the Radiation Therapy Oncology Group. The registry lists an enrollment of 60 and identifies the study status as completed. The recorded start was 2004-09 and the primary completion was 2012-04. Allocation is listed as not applicable, and masking as none. There is no randomized comparison described, so the design does not provide a concurrent group treated under an alternative regimen.

The treatment strategy combined induction paclitaxel and carboplatin with concurrent radiation therapy, then surgery, followed by consolidative chemotherapy. The interventions listed in the record include induction carboplatin, induction paclitaxel, radiation therapy, resection, consolidation carboplatin, and consolidation paclitaxel. The available arm description gives some regimen details but ends before the full radiation description is complete. Details beyond the text in the registry record are not reported in the registry record.

The primary endpoint was mediastinal nodal clearance, assessed at completion of concurrent chemotherapy and radiation therapy, up to 14 weeks. Its stated population was eligible patients with adequate pathologic information about mediastinal nodal status. The record describes a two-stage decision design, with thresholds based on evaluable participants. This is a planned monitoring and decision structure, not a report of the number actually evaluable or of the observed endpoint.

Other listed outcomes include pathological response, surgical morbidity, ability to undergo resection, resection status, survival, progression-free survival, and adverse-event grade. The registry record does not provide results for these outcomes in the registry record for this analysis. It also does not report the full protocol, the rationale for the enrollment target, interim review procedures, or a complete statistical analysis plan. A completed status and an enrollment entry describe the registry record; by themselves they do not establish the result of the primary endpoint.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Eligible patients who had adequate pathologic information of mediastinal nodal status.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: All eligible patients who started protocol treatment.

6. Primary Endpoint

The registry lists the primary outcome as Mediastinal Nodal Clearance Rate, measured over At completion of concurrent chemotherapy and radiation therapy, up to 14 weeks..

The primary endpoint is the mediastinal nodal clearance rate after concurrent chemotherapy and radiation therapy. It concerns whether mediastinal nodal disease is found to have cleared, based on pathologic information, among eligible participants whose nodal status can be assessed adequately. The registry gives the assessment time as completion of concurrent treatment, up to 14 weeks. It does not provide a more detailed operational definition of clearance or specify how ambiguous or incomplete pathology findings were classified.

The endpoint is expressed as a count of participants with clearance, and the endpoint description frames that count as evidence about a rate. A binary outcome is useful in a phase II setting because each evaluable participant can be classified according to whether the specified event occurred. Counts can then be assessed against a decision rule. Here, the record states that meeting both stage thresholds would support a conclusion of a 70% mediastinal nodal clearance rate compared with a 50% reference rate.

That reference is part of the design, not an observed comparison group. The record does not report the actual number with clearance, the endpoint denominator, or the resulting rate. It also does not report a confidence interval or other measure of precision for the observed endpoint. Consequently, the endpoint’s clinical role and planned decision criteria can be described, but its result cannot be interpreted from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary participant-level outcome: mediastinal nodal clearance is either recorded or not recorded for an evaluable participant under the trial’s endpoint definition. The registry expresses the endpoint as a number of participants and describes a rate-based conclusion. In general, binary outcomes are analyzed by counting participants in each outcome category and relating the event count to the relevant analysis population. The registry record does not report the endpoint count or denominator, so no observed rate can be stated.

What does the two-stage design do?

The primary endpoint description names Simon’s two-stage design. In general, this type of design specifies an early decision point and a later decision point, with a rule based on the accumulating outcome counts. It can allow a study to stop according to prespecified evidence rules rather than simply continuing to its planned maximum in every circumstance. The registry’s stated criteria require at least 12 of the first 21 evaluable participants and at least 27 of the first 45 evaluable participants to have mediastinal nodal clearance for the specified conclusion. The record gives the design’s stated power as 90% and its type I error as 10%.

Those operating characteristics describe the design under its assumptions; they are not evidence that the observed endpoint met the rule. They also do not quantify the precision of a result that was actually observed. The registry record does not report whether the study used the planned stages exactly as stated, whether any stopping decision occurred, or the evaluable counts at the decision points. Those details are not reported in the registry record.

What do the design reference and target mean?

The 50% reference rate and the 70% conclusion in the endpoint description define the contrast the design was built to assess. They are not rates observed in two groups. A single-group study can compare its outcome with a prespecified reference, but this is different from directly assigning participants to a treatment group and a control group. A reference-rate analysis depends on the suitability of the reference and the assumptions used to plan the decision rule. The registry record does not give the source or justification for the 50% reference, beyond stating it in the endpoint description.

In general, a type I error is the chance of making a positive design conclusion under the reference condition when that conclusion should not be made, according to the model used for planning. Power is the chance of making the intended conclusion under a specified alternative condition, also according to that model. The registry states 10% type I error and 90% power for the design. These are planned operating characteristics, not an observed power calculation for the completed study. The registry does not provide the assumptions or full calculation details needed to independently assess those properties.

How should uncertainty be understood?

A confidence interval is a statistical range intended to convey uncertainty around an estimated quantity under a specified method. In general, a wider interval indicates less precision, while a narrower interval indicates greater precision. For a binary endpoint, the amount of information depends on both the number of observed events and the number of participants with usable endpoint information. The primary endpoint’s observed count, denominator, and confidence interval are not reported in the registry record. It is therefore not possible to describe the endpoint’s numerical precision from this record.

The registry’s stage thresholds are decision criteria; they are not confidence limits. Meeting a threshold would not mean that the true rate is known exactly, and falling short would not establish that the rate equals the reference. Statistical decisions are made under uncertainty and depend on the prespecified model and rules. The endpoint description does not report a confidence interval method or an estimate of the observed clearance rate.

What do the planning illustrations represent?

The sample-size planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects expressed as a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power. The stated assumptions describe hypothetical two-arm planning, whereas this trial is registered as single-group. They should not be read as the actual design of this study or as evidence that a two-arm comparison took place. The illustration values and the observed effect needed to interpret them are not reported in the registry record.

More generally, planning calculations depend on the chosen outcome, target effect, variability or event rate, significance level, allocation, and analysis method. A change in those assumptions can change the resulting sample size. Over-dispersion, in general, means that outcome variation is greater than a simple statistical model assumes; if present and relevant, it can make nominal uncertainty estimates too optimistic unless accounted for. The registry record does not report an over-dispersion assessment, and there is no basis to claim it affected this trial.

What analysis details remain unavailable?

The registry record does not report an analysis method for the primary endpoint beyond the described decision design. It does not report whether estimates were adjusted, how any protocol deviations were handled, or whether analyses followed an intention-to-treat or evaluable-participant approach beyond the endpoint population wording. It also does not report a statistical analysis plan, an interim-analysis report, or the actual endpoint results. Without those details, the design can be explained, but the execution and numerical findings cannot be independently reconstructed.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Chemoradiation, Surgery, Chemotherapy")
n      <- c(43)
events <- c(27)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Chemoradiation, Surgery, Chemotherapy  43  27  62.8  46.7  77.0  47.9  75.6  47.8  75.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures, including the primary mediastinal nodal clearance endpoint and outcomes concerning pathological response, surgical morbidity, resection, survival, progression-free survival, and adverse-event grade. Assessing several outcomes raises a general multiplicity issue: when many statistical questions are tested, the chance of at least one apparently notable result can increase unless the analysis plan accounts for the number and relationship of those questions. The registry record does not report a multiplicity adjustment, a hierarchy for interpreting secondary outcomes, or a plan distinguishing confirmatory from exploratory analyses.

The primary endpoint population is described as eligible participants with adequate pathologic information about mediastinal nodal status. That definition does not give the number of participants meeting the definition. The registry record does not report the endpoint denominator, the numbers excluded from the endpoint population, or how inadequate, unavailable, or indeterminate pathology information was handled. Missing-data methods, imputation, and sensitivity analyses are not reported in the registry record.

The record lists no analysis results. It does not describe an intention-to-treat analysis, a per-protocol analysis, or a complete-case rule for the primary endpoint. The words “evaluable patients” appear in the stage criteria, but the registry does not state how evaluability was determined in practice or how participants who did not reach a later stage were treated analytically. These distinctions matter because the denominator used for an endpoint can affect how its rate is interpreted.

There is also no reported statistical analysis plan establishing how the secondary outcomes were analyzed, how repeated outcome assessments were handled, or how time-to-event outcomes were censored. The absence of these details means that multiplicity, missingness, and analysis-population choices cannot be evaluated from the registry record. It does not establish that no such procedures existed; their details are not reported in the registry record.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hemoglobin (5.3%), nausea (5.3%), leukopenia nos (5.3%), pneumonitis nos (5.3%), hypotension nos (5.3%).

The registry lists a single treatment group for the adverse-event information. For serious adverse events, 57 participants are listed at risk and 19 are listed as affected. For other adverse events, the corresponding counts are 57 at risk and 57 affected. These are participant counts in the registry record; they should not be mistaken for event counts, since one participant could potentially experience more than one event. The record does not provide further detail on that distinction.

The serious adverse-event terms with the largest listed percentages are Hemoglobin, Nausea, Leukopenia NOS, Pneumonitis NOS, and Hypotension NOS, each at 5.3%, followed by Dysphagia at 3.5%. These are reported terms and percentages, not evidence that the treatment caused each event. The registry record does not provide comparative safety results from another group or enough detail to assess attribution, duration, recurrence, or severity distributions for each term.

Deaths are not reported in the registry record. The listed counts should be read descriptively and in the context of the limited information available, not as a complete characterization of the regimen’s safety.

13. What the Result Does, and Does Not, Mean

The record’s clearest statistical contribution is its description of a prospective single-group question and a decision rule for mediastinal nodal clearance. It specifies a 50% reference, a 70% conclusion under stated count thresholds, and the timing and population for the primary assessment. These details show what the study was designed to assess. They do not show whether the endpoint reached the target or what the observed clearance rate was.

The study is registered as phase II, completed, and with enrollment of 60. Those design and status entries do not substitute for a results report. The primary endpoint count, denominator, rate, and confidence interval are not reported in the registry record. Nor are results for the listed secondary outcomes reported in the registry record. The completion status should therefore not be interpreted as confirmation of a positive or negative endpoint finding.

Because there is no concurrent comparator arm, the registry information cannot establish how outcomes would have differed under another treatment strategy. The reference rate used in the design is not a group of participants observed under an alternative treatment. A single-group result, if available, would describe outcomes among the analyzed participants and could be assessed against the chosen reference, but it would not by itself establish a causal advantage over another regimen.

The safety counts offer a limited descriptive account of affected participants and listed event terms. They do not establish that the treatment caused the events, and the record does not report a comparative safety analysis. The figures also do not answer questions about how adverse events changed over time or how they related to treatment exposure.

Accordingly, the registry record permits a careful explanation of the study question and statistical design, but not a conclusion about treatment efficacy, comparative benefit, or the precision of an observed primary outcome. Those conclusions require reported outcome results and analysis details that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable for pairing a clinically specific pathology-based endpoint with a single-group phase II design. Mediastinal nodal clearance is assessed after concurrent chemotherapy and radiation therapy, before the later surgical and consolidative components of the treatment sequence. The design’s decision rule makes explicit how endpoint counts would be used to support a conclusion about a target rate relative to a reference. That makes the study question more concrete than an endpoint label alone.

The two-stage structure also illustrates how a study can define decision points in advance. In general, such a design can set rules for evaluating accumulating evidence while controlling specified error properties under assumptions. The registry reports the thresholds and the stated design characteristics, but does not report whether the criteria were met or how the study proceeded through its stages. It therefore illustrates the logic of the planned method without supplying evidence about its realized result.

The single-group structure is equally important. A planned comparison with a reference rate is not the same as a randomized comparison with a concurrent control. Readers must keep those questions separate: a study may assess whether an outcome is consistent with a prespecified target, while still not estimating the difference that would arise under an alternative treatment. This distinction is central when interpreting phase II designs.

The record also shows why endpoint population definitions and outcome reporting matter. The primary assessment is restricted to eligible participants with adequate pathologic information, while the enrollment entry covers the registered study population. Without the actual evaluable count and result, readers cannot determine the endpoint denominator or its precision. The trial’s statistical design can be understood, but its primary findings cannot be inferred from design criteria alone.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main trial-specific limitation for interpreting the registry record is the absence of the primary endpoint result. Although the registry gives thresholds for a conclusion about mediastinal nodal clearance, it does not report the observed count, rate, denominator, or uncertainty. The criteria should therefore be read as planned rules, not as findings.

The single-group design has no concurrent randomized comparator. The 50% reference in the endpoint description is a design benchmark, not an observed control group. As a result, the design description alone cannot support a direct treatment comparison. It also cannot separate the contribution of the individual treatment components in the sequence of induction chemotherapy, radiation, surgery, and consolidative chemotherapy.

The endpoint depends on adequate pathologic information and on evaluability at the stated stages. The record does not report how many participants had adequate information, how evaluability was determined in practice, or how missing or indeterminate nodal assessments were handled. The listed enrollment of 60 does not resolve those questions. The endpoint’s detailed pathology definition is also not reported in the registry record.

The treatment description in the registry is incomplete, and the record does not provide a full protocol or complete statistical analysis plan. Detailed eligibility criteria, analysis populations, missing-data procedures, multiplicity handling, and outcome estimates are not reported in the registry record. The secondary outcomes are named, but their results are not available in the registry record used.

Safety interpretation is limited to the recorded counts and selected terms. Comparative safety, attribution, and fuller event-level details are not reported in the registry record. These limits restrict conclusions about both the primary endpoint and the broader balance of outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Paclitaxel, Carboplatin, and Radiation Therapy in Treating Patients Who Are Undergoing Surgery for Stage III Non-Small Cell Lung Cancer — NCT00096226.
• Registry record last updated 2019-09-18; status last verified 2019-09.
• Results first posted 2014-02-27.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00096226 is a completed phase II, single-group trial sponsored by the Radiation Therapy Oncology Group. It lists an enrollment of 60 and concerns patients undergoing surgery for stage III non-small cell lung cancer. The registered treatment sequence includes induction paclitaxel and carboplatin with concurrent radiation therapy, followed by resection and consolidative chemotherapy. The recorded study start was 2004-09, with primary completion at 2012-04.

The primary endpoint is mediastinal nodal clearance, assessed at completion of concurrent treatment, up to 14 weeks, among eligible participants with adequate pathologic information. The registry describes Simon’s two-stage design and its stated count thresholds, but does not report the observed primary endpoint result. It also lists secondary outcomes and selected adverse-event counts. Detailed outcome estimates, analysis methods, and missing-data procedures are not reported in the registry record.